Mixedbread

Quickstart

Build your own AI Search Engine in just a few minutes. This quickstart will show you how to create Vector Stores that automatically understand and index any data format from PDFs, images to code.

Install the SDK

Install the SDK for your preferred language:

pip install mixedbread

Get your API key

  1. Go to the page.
  2. Save your key to your .env file:
MXBAI_API_KEY=your-api-key

Build Your AI Search Engine

Let's create a Vector Store that automatically processes and understands your files, making them searchable with natural language queries.

1. Create a Vector Store

First, create a Vector Store, your AI-powered search index that will understand and organize your data.

2. Upload and Process Files

Upload your files to the Vector Store. Vector Stores automatically process any format: PDFs, images, documents - without requiring separate parsing tools.

3. Search Your Data

Now you can perform search through your files using natural language queries:

Inference Models

Beyond Vector Stores, you can also use our embedding and reranking models directly for custom applications.

Create Embeddings

Generate semantic vectors from text for similarity search, clustering, or custom retrieval systems:

Rerank Search Results

Improve search quality by reranking results based on semantic relevance:

Tools

Enhance your workflow with our powerful tools for easier integration and development.

CLI

Use our CLI to manage Vector Stores directly from your terminal:

# Install the CLI
npm install -g @mixedbread/cli

# Create a Vector Store
mxbai vs create "my-knowledge-base"

# Upload files
mxbai vs upload "my-knowledge-base" ./documents/

# Search from terminal
mxbai vs search "my-knowledge-base" "How does authentication work?"

MCP (Model Context Protocol)

Enable your LLM to search your Vector Stores directly with our hosted MCP server - no installation required!

Configure it with your AI assistant:

Now your AI assistant can search your Vector Stores automatically during conversations.

Next Steps

  • : Deep dive into Vector Store capabilities
  • : Learn about our embedding and reranking models
  • : Complete CLI documentation
  • : Set up Model Context Protocol integration
  • : Complete API documentation
Overview
The AI Search Engine that makes all your unstructured data understandable and usable for AI. Build intelligent search experiences with Vector Stores.
Vector Stores Overview
The AI Search Engine that makes all your unstructured data understandable and usable for AI. Provide AI with the right context at the right time.

Last updated: July 30, 2025